prompt-for-others
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest local project context to generate prompts for other agents to execute. While this creates a potential surface for instructions to be passed between agents, the skill includes several defensive guidelines for the AI.
- Ingestion points: The agent reads environment context and fix details from the local file system (SKILL.md).
- Boundary markers: The skill requires the final output to be wrapped in a single blockquote, providing a clear boundary for the generated content.
- Capability inventory: The skill itself has no active tool capabilities enabled, as it uses the
disable-model-invocation: trueflag. The generated prompts describe standard development tools such as Homebrew (brew) and system services (launchctl). - Sanitization: Instructions explicitly forbid the inclusion of secrets, tokens, or private URLs, and mandate the use of 'stop rules' to prevent unintended installations or destructive actions.
Audit Metadata